ER alpha selective chromone, isoxazolylchromones, induces ROS-mediated cell death without autophagy

Article abstract of DOI:10.1111/cbdd.13510

Chromones are recognized as privileged structures and useful templates for the design of novel compounds with promising pharmacological activity. Several reports implicate chromone scaffold as an antitumor agent. The present study highlights synthesis, do

Full text of DOI:10.1111/cbdd.13510

Received: 1 October 2018
DOI: 10.1111/cbdd.13510
Revised: 11 January 2019
Accepted: 9 February 2019
|
|
R E S E A R C H A R T I C L E
ER alpha selective chromone, isoxazolylchromones, induces ROS‐
mediated cell death without autophagy
Swati Kaushik1
Mydhily Nair1
,2
1,3
2
Asha Lekshmi1
Thankayyan Retnabai Santhoshkumar
Leena Chandrasekhar
1
|
Rahul Sanawar
Seema Bhatnagar
|
|
|
1
|
|
1Cancer Research Program 1, Rajiv Gandhi
Centre for Biotechnology, KINFRA
Campus, Trivandrum, Kerala, India
Abstract
Chromones are recognized as privileged structures and useful templates for the de-
sign of novel compounds with promising pharmacological activity. Several reports
implicate chromone scaffold as an antitumor agent. The present study highlights syn-
thesis, docking, and potential activity of isoxazolylchromones, 3(a–f), a new class of
compounds as potential agents exhibiting ERα antagonism and ERβ agonism.
Molecular docking studies determined the binding site of compounds 3(a–f) in ERα
and ERβ. All the analogues synthesized showed preferential cytotoxicity in ERα+
cell line (MCF‐7) compared to ERα‐ cell line (MDA‐MB‐231). Among the ana-
logues synthesized, analogue 3d exhibited increased cytotoxicity. ERα silencing ex-
periments confirmed the ERα selective nature of ligands. Transactivation assay on
compound 3d indicated the down‐regulation of ERα luciferase reporter gene expres-
sion and induction of ERβ GFP in the treated cells. Cell cycle analysis revealed an
increase in sub-G0/G1 population on treatment with analogue 3d as compared to
control. Similar to tamoxifen, 3d‐induced cell death is mediated through an increase
in ROS as evidenced by change in roGFP ratio. Interestingly, the compound 3d in-
duced mitochondrial trans‐membrane potential loss and caspase activation without
indication of autophagy compared to tamoxifen that induced autophagy in the treated
cells. Lack of significant autophagy and induction of ERβ signaling by the new com-
pound place them as a better ERα antagonist.
2
Novel Molecule Synthesis
Laboratory, Amity Institute of
Biotechnology, Amity University, Noida,
Uttar Pradesh, India
3
Manipal Academy of Higher Education
(
MAHE), Manipal, Karnataka, India
Correspondence
Thankayyan Retnabai Santhoshkumar,
Cancer Research Program‐1, Rajiv Gandhi
Centre for Biotechnology, KINFRA
Campus, Trivandrum, Kerala, India.
Email: trsanthosh@rgcb.res.in
and
Seema Bhatnagar, Amity Institute of
Biotechnology, Amity University, Noida,
Uttar Pradesh, India.
Email: sbhatnagar1@amity.edu
Funding information
DST | Science and Engineering Research
Board (SERB), Grant/Award Number:
PDF/2016/004001
K E Y W O R D S
autophagy, Caspase, cell cycle, estrogen receptor, isoxazolylchromones, ROS, SERM, transactivation
1
|
INTRODUCTION
response. There is compelling evidence to indicate that el-
evated levels of the female sex hormone estrogen lead to a
Estrogen receptor (ER) signaling is exceedingly complex and
plays an essential role in both normal physiology and diverse
pathological conditions including breast cancer. Although ER
acts as ligand‐activated transcription factors, the ER response
is complex and cannot be explained by “classical concept”
of receptor activation or deactivation upon ligand binding.
A complex interplay of signaling cross talk comprising “ge-
nomic” and “non‐genomic” pathways leads to transcriptional
predisposition toward breast cancer (Henderson & Feigelson,
2000). As per estimates, nearly 70% of breast cancers express
ER (Masood, 1992; Nilsson et al., 2001), and in these cases,
estrogen is considered to promote tumor growth. Estrogen
mediates its cellular signaling through two ER subtypes ER
alpha (ERα) and ER beta (ERβ); both are members of the
nuclear receptor (NR) superfamily (Heldring et al., 2007).
Differential tissue distribution of ERα and ERβ adds to the
Chem Biol Drug Des. 2019;1–16.
wileyonlinelibrary.com/journal/cbdd
© 2019 John Wiley & Sons A/S
|
1

2
KAUSHIK et Al.
|
complexity of the problem. The alpha subtype has a more
prominent role on the mammary gland and uterus whereas
the beta subtype seems to have the more profound effect on
the central nervous and immune system. In addition, ER beta
signaling generally counteracts the ERα promoted cell hy-
perproliferation in tissues such as breast and uterus (Paterni,
Granchi, Katzenellenbogen, & Minutolo, 2014). The estro-
gen receptor ligand‐binding domain (LBD) has been the pri-
mary target for nuclear hormone receptor drug discovery. A
comparison of LBD of ERα and ERβ indicates that they share
a high degree of similarity in residues that line the binding
cavity (Kuiper et al., 1997) while ER subtype selectivity is
ultimately determined by structural differences in the LBDs
of ERα and ERβ. Structural modeling by Minutolo and co‐
workers reported that residues in ERα and ERβ in contact
with ligand differ at only positions: In helix 5, Leu384 of
ERα corresponds to Met336 of ERβ and Met421 of ERα cor-
responds to Ile373 of ERβ in loop 6‐7 (Minutolo, Macchia,
Katzenellenbogen, & Katzenellenbogen, 2011). This differ-
ence confers variations in the total volume of ER isoforms
chemical entities that may tackle this challenge. A literature
search on bioactive natural metabolites indicates that chro-
mone scaffold possess a wide spectrum of biological activity
(Pick et al., 2011). Most prominently the chromone scaf-
fold has been explored for anticancer properties (Middleton,
Kandaswami, & Theoharides, 2000; Momoi et al., 2005). In
the current study, we have aimed to identify ER subtype‐se-
lective ligands from several new isoxazolylchromones 3(a–f).
Docking studies, dose–response analysis, gene silencing ex-
periments, transactivation assay, GFP‐labeled ERβ assay, cell
cycle analysis, apoptotic studies, ROS induction, and auto-
phagic analysis have confirmed the new ligands as a potential
ERα selective antagonist. The studies also emphasize that
the lead molecule elicits less cell survival signaling and in-
creased ERβ agonist activity.
2
2
METHODS AND MATERIALS
Cell culture and gene silencing
|
.1
|
experiments
3
LBD. The overall volume of the ligand‐binding site is 490 Å
3
in ERα and 390 Å in ERβ (Ottow & Weinmann, 2008). These
MCF‐7 cells were grown in monolayer culture in RPMI‐1640
(Sigma Chemicals Co., USA) supplemented with 10% heat‐
inactivated fetal bovine serum (FBS) and antibiotics in a
differences can contribute to ligand selectivity on account of
ring substitution, stereochemistry, and conformational orien-
tation. Literature also indicates that the anchorage of a ligand
into the receptor LBD is achieved by various interactions (H‐
bond, Van der Waals interaction) which most probably confer
a broad spectrum of conformations leading to the recruitment
of different sets of co‐regulators (Laïos et al., 2007). Reports
also suggest that ligands promote or prevent co‐activator
binding based on the shape of the estrogen or anti‐estrogen
receptor complex (Shiau et al., 1998).
humidified atmosphere of 5% CO at 37°C. For silencing
2
experiment, cells were seeded in the 6‐well plates at desired
densities. (a) Gene silencing using ERα siRNA on MCF‐7
Cells: MCF‐7 cells were transfected with ERα siRNA
(Santa Cruz, #SC 29305) using oligofectamine (Invitrogen,
#12252‐011) according to manufacturer's protocol. (b) Drug
Addition: For drug treatment, after overnight culture of cells
in 6‐well plates, the medium was replaced with fresh DMEM
containing 5% FBS and the compounds to be tested. For im-
aging, 5% FBS containing phenol red‐free DMEM was used.
Concentrated stocks of compound 3d and 4HT were prepared
inDMSOandstoredat−80°C. (c)LiveCellStaining–Hoechst
33342 staining: Hoechst 33342 is a popular cell‐permeant
nuclear stain that emits blue fluorescence when bound to
double‐strand DNA at A‐T rich regions. Following treatment
with compound 3d at different concentrations and known
drug 4HT, ERα gene silenced and non‐silenced MCF‐7
cells were stained with 2 μg/ml Hoechst 33342 (Molecular
Probes #H1399) in phenol red‐free DMEM containing 5%
FBS by incubating at 37°C for 10 min. After washing twice
with PBS, Hoechst 33342 fluorescence intensity was imaged
for detecting pyknotic nuclei using DAPI filter set under the
fluorescent microscope (Nikon TiE). Images were analyzed
using NIS elements software. (d) SDS–PAGE and Western
blotting: The MCF‐7 cells treated with (ERα siRNA) or with
vector control siRNA were washed with PBS and lysed using
phosphorolysis buffer supplemented with protease inhibitors.
After lysis, the suspension was centrifuged at 12,000 rpm for
20 minutes and the supernatant containing the whole cell
Therapeutic agents that target the ER‐positive cancers
are referred to as selective estrogen receptor modulators
(
SERMs). Thus far, only five non‐steroidal SERM and one
steroidal anti‐estrogen have been marketed (Maximov, Lee,
Jordan, 2013). For almost three decades, prototypical
&
SERM, tamoxifen has been the drug of choice for the first‐
line therapy in both early and advanced ER‐positive breast
cancer (Peng, Sengupta, & Jordan, 2009). Unfortunately,
long‐term use of tamoxifen has been linked with undesir-
able side‐effects and acquired resistance (Beato, Herrlich,
&
Schütz, 1995; Chang, Kim, Malla, & Kim, 2011; Farhat,
Lavigne, & Ramwell, 1996; Turner, Riggs, & Spelsberg,
994). Even though diverse mechanisms of hormone resis-
1
tance are reported; induction of various survival signaling
including autophagy is a major concern. It has been reported
that tamoxifen and most of the currently used SERM are
capable of eliciting protective autophagy in the target cells
(
Hongyi et al., 2017; Sommer & Fuqua, 2001).
In this regard, efforts to discover and develop new and
more specific subtype‐selective ligands, whether they are
agonist or antagonist, have re‐energized search for new

KAUSHIK et Al.
3
|
proteins was immediately estimated for protein, denatured,
and resolved by SDS–PAGE. The protein concentration for
each sample was estimated using Bradford's method. Protein
samples were denatured in sample buffer containing β‐mer-
captoethanol and SDS at 100°C for 8 min and resolved in a
each protein lysate were normalized to the Renilla lucif-
erase readings. Estradiol was used at the concentration of
0.0005 μg/ml, and 4HT was used at a final concentration
of 5.63 μg/ml.
1
0% acrylamide gel. Equal amount of protein from the whole
2
.3
ERβ‐GFP reporter assay
|
cell lysates of MCF‐7 cells transfected with control siRNA
and ERα siRNA was loaded and electrophoretically run
along with prestained protein marker at a constant voltage
of 80 V through the stacking gel and at 100 V through the
separating gel. The resolved bands in the SDS–PAGE were
electroblotted onto nitrocellulose membrane (Hybond‐C
pure, Amersham) using Bio‐Rad Mini‐PROTEAN III wet
blot apparatus at 70 V for 2 hr in ice. After the transfer, the
blots were washed with TBST buffer for 10 mins. The blot-
ted membrane was blocked with 10% non‐fat milk in Tris‐
buffered saline (TBS) containing 0.2% Tween‐20 for 1 hr at
room temperature to avoid the non‐specific binding. Specific
proteins were detected by incubating overnight with appro-
priate primary antibody in TBST buffer containing 5% BSA
at 4°C followed by incubation with secondary antibody cou-
pled with alkaline phosphatase or horse‐radish peroxidase
conjugates. (The primary antibodies were mouse anti‐ERα
ERβ‐EGFP expressing MCF‐7 cells were generated using
FuGENE HD Transfection reagent according to manufacturer's
®
protocol (Promega, Madison, WI 53711 USA); pEGFP‐C1‐ER
beta plasmid was purchased from Addgene (plasmid #28237)
(Addgene, Cambridge, MA 02139, USA). Homogeneously ex-
pressing single‐cell colonies were selected using 800 μg/ml G418
antibiotic selection (Invitrogen, Carlsbad, CA) followed by FACS
sorting. For live cell imaging, the ERβ‐EGFP cells were seeded
on a chambered cover glass (Lab‐Tek, Nunc, Rochester, NY,
USA) and grown in DMEM containing 10% FBS. Once the cells
reached a confluency of 80%, the spent media was replaced with
phenol red‐free DMEM supplemented with 5% charcoal‐treated
FBS to reduce the contaminating steroids from the serum. After
4 hr, cells were treated with compound 3d (32 μg/ml), estradiol
(0.0005 μg/ml), and 4HT (5.63 μg/ml). Imaging was carried out
using a Leica SP8WLL laser scanning confocal microscope. The
images were captured using the 20X objective, and the emission
was collected at 500‐550 nm after exciting at 488 nm.
(
#SC‐8002), mouse anti‐β‐actin (#A1978) purchased from
Santa Cruz, USA and Sigma Chemicals Co., USA; used at
1
:350 and 1:1,000 dilution, respectively. The secondary an-
tibodies used were horse‐radish peroxidase (HRP) labeled
anti‐mouse secondary antibodies (Sigma Chemicals Co.,
USA) at 1:4,000 dilution). The bands were developed with
2
.4
ROS analysis using roGFP cells
|
Breast cancer cells were transfected with plasmid encoding for
redox‐sensitive GFP, roGFP2 using FuGENE HD transfec-
®
3
,3′‐diaminobenzidine (DAB/H O ) in 50 mM Tris, pH 7.6
2
2
or using enhanced chemiluminescence as per the manufac-
turer's protocol.
tion reagent according to manufacturer's protocol (Promega,
Madison, WI, USA). The stable cells expressing roGFP were
developed after G418 selection for 2 months. For the analysis of
roGFP by flow cytometry, cells were seeded on 24‐well plates
and treated with 4HT (4HT‐1 = 5.63 μg/ml; 4HT‐2 = 11.26 μg/
ml) and 3d (3d‐1 = 32 μg/ml and 3d‐2 = 64 μg/ml). The cells
were trypsinized and analyzed using FACS AriaIII equipped
with 405 and 488 nm laser lines. The emission was collected
using 530/30 filter from the two laser paths in ratio mode. The
cell population with increased 405/488 ratio from untreated
cells were gated to calculate the percentage of cells with an in-
crease in ROS. For microscopic imaging of roGFP, cells were
grown on glass bottom plates and treated as above. The emis-
sion was collected at 515 ± 15 nm after exciting using 405 and
488 filter sets in ratio mode using NIS element software utiliz-
ing fully motorized fluorescent microscope (Nikon TiE).
2
.2
Transfection and Luciferase reporter
|
gene assay
MCF‐7 cells after attaining 60%–70% confluency, 10%
FBS containing DMEM media supplemented with antibi-
otic was removed and replaced with 5% charcoal‐treated
FBS containing media to reduce the contaminating ster-
oids from the serum. After 4 hr, cells were transfected with
pGL3‐(ERE)‐Luc plasmid (400 ng) using Lipofectamine
LTX and Plus reagent kit (Invitrogen.Cat.15338100)
®
™
according to manufacturer's protocol. To evaluate the ef-
ficiency, 100 ng of SV40‐Renilla luciferase (pRL‐SV40)
vector (Promega Madison, USA) was used as internal con-
trol for normalization. The plate was then incubated for
another 20 hr in the humidified 5% CO incubator. After
2
2
.5
Autophagy analysis using MCF‐7 LC3
|
2
0 hr of transfection, cells were subjected to media con-
GFP cells
taining either 3d or 4HT separately and the plate was then
kept back for another 24 hr. After 24 hr, luciferase assay
was performed (according to the protocol of kit provided
by Promega Madison, USA). The luciferase readings for
MCF‐7 cells were transfected with plasmid encoding LC3 GFP
®
using FuGENE HD transfection reagent according to manu-
facturer's protocol (Promega, Madison, WI 53711 USA). The

4
KAUSHIK et Al.
|
FIGURE 1 Illustrates scheme for the synthesis of isoxazolylchromones 3(a–f)
cells were stained with TMRM and then were exposed to 3d at
was done using BDAttoVision™ (version 1.6/435, Becton
Dickinson Biosciences, USA) by applying simple polygon
segmentation, for nuclear regions identification using thresh-
old level with FRET ratio loss (increased DEVDase activity)
corresponding to increased ECFP/EYFP ratio signal. The re-
sults of FRET loss corresponding to increased ECFP/EYFP
ratio were used to identify cells with caspase activation.
Based on this, the percentage of cells with caspase activation
was scored for each treatment as compared with control.
its IC (32 μg/ml) value and 4HT at 2.81 μg/ml for 24 hr. The
50
autophagy‐specific aggregation of LC3 GFP was analyzed using
GFP and PE filter sets under the fluorescent microscope, and the
images were analyzed using NIS element software. Cells with
more than 15 GFP punctae were scored as autophagic.
2
.6
Cell cycle analysis by propidium
|
iodide stain
4
MCF‐7 cells seeded in 60‐mm dishes (2 × 10 cells/ml) over-
night were treated with three concentrations of 4HT (5 μg/ml,
2
.8
Molecular docking protocol
|
1
0 μg/ml, and 20 μg/ml) and analogue 3d (15, 30, 60 μg/ml)
Protein preparation: The ER PDB (http://www.pdb.
org) IDs retrieved were 3ERT (Shiau et al., 1998), 2QTU
(Richardson et al., 2007), and 2FSZ (Wang et al., 2006), re-
spectively. The crystal structure was prepared with the help
of the protein preparation wizard from the workflow option
of the Maestro 9.3 Schrodinger suite. The water molecules
were deleted, hydrogens were added, bond orders were as-
signed, and N‐acetyl and N‐methyl amide capping groups
were added to the N‐terminus and C‐terminus, respectively.
Finally, a restrained minimization of the receptor model was
performed setting the default constraint of RMSD of 0.30
Ǻ and OPLS 2005 force field. The first minimization was
performed constraining the heavy atoms in order to allow
free rotation of the hydrogen. Subsequently, minimizations
were performed by decreasing the constraints on the heavy
atoms. Receptor grid preparation: The 3D structure of ER
was prepared using all three PDB entries and was used
to generate Glide scoring grid for the successive docking
for 48 hr. Cells were trypsinized, washed twice with ice‐cold
PBS, and fixed with ice‐cold 70% ethanol. The cell pellet was
stained with propidium iodide for cell cycle analysis after
RNAase treatment and analyzed using BD FACS Aria III.
The single‐cell population identified were gated for sub‐G0
apoptotic cells using FACS Diva 7.0 software and analyzed.
2
.7
Caspase activation analysis
|
(
a) Image acquisition: The breast cancer cells (MCF‐7
SCAT3 NLS) expressing a FRET‐based caspase sensor were
described earlier (Joseph, Seervi, Sobhan, & Santhoshkumar,
2
011). For the detection of caspase activation, the stable cells
were seeded in 96‐well glass bottom plates at the desired den-
sity. After overnight culture, the medium was removed and
replenished with phenol red‐free DMEM containing 5% FBS
containing the known drug 4HT (5.63 μg/ml) as a positive
control and compound 3d to be tested at its IC value. Plates
3
calculations. The receptor grid box of size (10‐10‐10 Å )
5
0
™
were imaged under BD Pathway 435 Bioimager (Becton
Dickinson Biosciences, USA). The excitation/emission filter
for ECFP and EYFP FRET filter was used as described earlier
was generated with the centroid of the active site residues.
Default parameters were used, and no constraints were in-
cluded during grid generation. Ligand preparation: The
3D structures of the ligands were generated using Maestro,
and the geometry of ligands was optimized by molecular
mechanics using IMPACT (Banks et al., 2005) in a dy-
namic environment using standard TIP4P water model.
(
Joseph et al., 2011). Image for each well was acquired in the
respective channels using a dry 20X objective with NA 0.7.
Images were captured as 2 × 2 montage. (b) Postacquisition
segmentation and analysis: Postacquisition image analysis

KAUSHIK et Al.
5
|

6
KAUSHIK et Al.
|
FIGURE 2 (a) Illustrates the ligand
interaction map of 3d with residues in
ERT and its interactions with key residues
3
Arg394 and Glu353. Also shown here is
the ribbon model representation of the
three‐dimensional structure of ERα (b) and
ERβ (c) docked with compound 3d shown in
green color, and superimposition with 4HT
is depicted in red color
The conformational models of the ligands were generated
using ligand preparation wizard (LigPrep.) Software tools
LigPrep2.2 and Epik1.6 were used to compute the prob-
able protonation states at pH 7.0 ± 3.0, all tautomers and
up to 32 stereoisomers for compounds. The energy mini-
mization was done using Optimized Potentials for Liquid
Simulations 2005 (OPLS 2005) force field. Energy mini-
mization was done using Polak–Ribiere conjugate gradient
and Truncated Newton conjugate gradient algorithms. The
convergence threshold of rms gradient of 0.01 was used.
Conformational models of the ligands were generated using
LigPrep. The lowest energy conformations of the ligands
were used for docking with the receptor model. Docking
and scoring protocols: The docking studies were carried
out between all the model of ER and the different conforma-
tions of the ligands. All docking calculations were carried
out using extra precision (XP) in built‐in Glide (Grid‐based
Ligand Docking with Energetics). For the flexible dock-
ing mode, a set of conformers for each input ligand were
generated by Glide after which an exhaustive search for
possible positions and orientations of ligand over the ac-
tive site were performed. The ligand poses generated by
Glide are then passed through a series of hierarchical filters
that evaluate the interaction of the ligand with the receptor.
The OPLS‐2005 force field was used for this purpose. A
small number of surviving docking solutions can then be
subjected to a Monte Carlo procedure to try and minimize
the energy score. The final energy evaluation was done
with Glide Score, and a single best pose was generated as
the output for a particular ligand. These minimized poses
are rescored using Glide Score (G‐Score) scoring function.
Model energy score (E model) was selected as the final
scoring function to select the best‐docked structure for each
ligand. E model combines Glide Score with the non‐bonded
interaction energy between the ligand and the protein, the
energy grid score, and (for flexible docking) the ligand
strain energy. The different conformations of the known
ligands were docked, and 1,000 poses per compound were

KAUSHIK et Al.
7
|
FIGURE 3 Dose‐dependent studies of isoxazolylchromones 3(a–f) plotted against % change in optical density at 570 nm in MCF‐7, MDA‐
MB‐231, and MRC‐5 cell lines via MTT assay. The data are means and SEM from three samples of each group. Two‐way ANOVA followed by
Bonferroni post‐test where p‐value <0.05 is significant for compounds
obtained. The analysis of the poses, complexes, and the
binding affinities between the receptor and ligands were
analyzed using Schrodinger's software suite.
(2.2 mmol) was added and the mixture was stirred at room
temperature. After 5 hr of reaction, TLC was monitored
and the product was extracted with EtOAc. The extract was
washed with water (3 × 10 ml), dried (Na SO ), and concen-
2
4
trated. The crude was subjected to column chromatography
2
.9
Statistical analysis
|
over SiO (60–120 mesh) using hexane: ethyl acetate 20% as
2
The experimental results were expressed as mean ± stand-
ard error of the mean of four replicates. Where applicable,
the data were subjected to one‐way and two‐way analysis of
variance (ANOVA) followed by Bonferroni post‐test using
GraphPad Prism (version 5.1). p‐value of ≤0.05 was re-
garded as significant.
eluent to give the desired isoxazolyl derivatives 3(a–f).The
spectral and analytical data of isoxazolylchromones are given
below.
3a. 3‐hydroxy‐2‐[3′,5′‐diphenylisoxazol‐4′‐yl]‐4‐chro-
mone obtained as semisolid from 2‐vinylchromone (2a)
−
1
and benzaldoxime (1a), IR (KBR, cm ): 3444 (OH), 1748
1
(
C=O),1687 (C=N); H NMR (500 MHz, DMSO, rt): δ
6
.91‐8.08 (14H,ArH), 9.79 (s,1H,OH); ESI‐MS m/z: 382.38
3
3
EXPERIMENTAL
|
+
(
M + 1).
3
b. 3‐hydroxy‐2‐[3′‐(4‐hydroxyphenyl)‐5′‐phenylisox-
.1
General procedure for the synthesis of
|
azol‐4′‐yl]‐4‐chromone obtained as semisolid from 2‐vinyl-
Benzaldoxime (1a–b)
−1
chromone (2a) and benzaldoxime (1b), IR (KBR, cm ):
1
Synthesis of benzaldoxime was performed according to the re-
ported procedure (Das, Mahender, Holla, & Banerjee, 2005).
3460(OH), 1795 (C=O), 1670 (C=N); H NMR (500 MHz,
DMSO, rt): δ 7.11‐8.74 (13H,ArH), 9.97 (s,2H,OH); ESI‐
+
MS m/z: 420.369 (M + Na).
3
c. 3‐hydroxy‐2‐[5′‐(2‐methoxyphenyl)‐3′‐phenylisox-
3
3
.2
General procedure for the synthesis of
|
azol‐4′‐yl]‐4‐chromone obtained as semisolid from 2‐vi-
‐hydroxy‐2‐[3′, 5′‐substituted isoxazole‐4′1‐
−
1
nylchromone (2b) and benzaldoxime (1a), IR (KBR, cm ):
yl]‐4‐chromone 2(a–f)
1
3
250 (OH),1758 (C=O), 1685 (C=N); H NMR (500 MHz,
A benzaldoxime (1a‐b, 1 mmol) and a 2‐vinylchromone
DMSO, rt): δ 4.04‐4.22 (m,3H,O CH ), 6.60‐8.89 (13H,
3
+
(
2a‐c, 1.2 mmol) were taken in acetonitrile (10 ml). CAN
ArH), 9.98 (s,1H,OH); ESI‐MS m/z: 411.40 (M) .

8
KAUSHIK et Al.
|
FIGURE 4 (a) Gene silencing
experiment. Expression of ERα protein
assessed by Western blot after transfecting
with ERα siRNA. The difference in the
expression of ERα in silenced as compared
to parent MCF‐7 cells is shown. (b) Effect
of 4HT and 3d on induction of apoptosis
in parental (cells transfected with control
siRNA) and silenced cells (cells transfected
with ERα siRNA). Parental and ERα
silenced MCF‐7 cells were assessed for
nuclear condensation following 48 hr
treatment with 4HT (10 μM) and 3d
(
15, 31.25, 62.5, 125, 250 μg/ml) by
Hoechst staining. (c) Data are expressed
as Mean±SD of nuclear condensation
expressed as a percentage (n = 3). Two‐way
ANOVA followed by Bonferroni post‐test
where p‐value <0.05 is significant. Scale
bar = 50 μm
3
d. 3‐hydroxy‐2‐[3′‐(4‐hydroxyphenyl)‐5′‐(2‐methoxyphe-
3f.
3‐hydroxy‐2‐[3′‐(4‐hydroxyphenyl)‐5′‐methylisox-
nyl)‐isoxazol‐4′‐yl]‐4‐chromone obtained as semisolid from 2‐
azol‐4′‐yl]‐4‐chromone obtained as semisolid from 2‐vinyl-
−1
−1
vinylchromone (2b) and benzaldoxime (1b), IR (KBR, cm ):
chromone (2c) and benzaldoxime (1b), IR (KBR, cm ):
1
1
3
420 (OH),1730 (C=O),1650 (C=N); H NMR (500 MHz,
3,350 (OH), 1,752 (C=O), 1691(C=N); H NMR (500 MHz,
DMSO, rt): δ 3.80‐3.93(m,3H,O CH ), 6.90‐7.84 (12H, ArH),
DMSO, rt):δ 3.30–4.09 (m, 3H, CH3), 6.87–8.46 (8H, ArH),
3
13
+
8
.48 (s,2H,OH); C NMR (300 MHz, DMSO, rt):δ 64.82
10.90 (s, 2H, OH); ESI‐MS m/z: 374.27 (M + K).
(
CH ), 112.64–143.55 (12CH), 112.64‐143.55 (6C), 156.27–
3
+
1
66.42 (5C), 182.13 (C=O); ESI‐MS m/z: 428.41(M + 1).
4
4
RESULTS
|
3
e. 3‐hydroxy‐2‐[5′‐methyl‐3′‐phenylisoxazol‐4′‐yl]‐4‐
chromone obtained as semisolid from 2‐vinylchromone (2c)
.1
Synthesis of isoxazolylchromones 3(a‐f)
−1
|
and benzaldoxime (1a), IR (KBR, cm ): 3,206 (OH), 1,769
1
(
(
C=O), 1,632 (C=N); H NMR (500 MHz, DMSO, rt):δ 2.49
A small library of isoxazolylchromones was synthesized
using 1, 3‐dipolar cycloaddition reactions by reaction be-
tween nitrile oxides and 2‐vinylchromones as dipolarophiles.
s, 3H, CH3), 6.66–7.88 (9H, ArH), 8.47 (s, 1H, OH); ESI‐
+
MS m/z: 319.31 (M) .

KAUSHIK et Al.
9
|
(
1a‐b). The resulting nitrile oxides were trapped by 2‐vi-
nylchromones (2a‐c). This method successfully resulted in
various derivatives of 3‐hydroxy‐2‐[3′, 5′‐substituted isox-
azole‐4′‐yl]‐4‐chromone (3a‐f, Scheme 1) in good yield.
The general reaction sequence involved in the synthesis of
3
(
‐hydroxy‐2‐[3′, 5′‐substituted isoxazole‐4′‐yl]‐4‐chromone
3a–f) is shown in Figure 1. The compounds were charac-
terized by using spectroscopic techniques. Detailed spectro-
scopic data are provided in Supporting information (Figures
S4–S19).
4
.2
Molecular docking studies
|
Docking was carried out using Maestro 9.6 Schrodinger
suite. The PDB entries chosen for the studies were 3ERT
(
Shiau et al., 1998), 2QTU (Richardson et al., 2007), and
2
FSZ (Wang et al., 2006). The PDB entry 3ERT chosen is
a complex of ERα with 4‐hydroxytamoxifen (4HT). The se-
lection of 2QTU entry was based on the structural similar-
ity of the synthesized ligands with benzopyranones. Native
ERβ PDB entry is not available; therefore, 2FSZ was selected
which is a complex of 4HT with ERβ. Computational tools
have been used to elucidate the binding site of compounds.
For this, binding affinities of the synthesized compounds with
the receptor were compared using ligand interaction map.
The docking of compounds has been studied at ERα,
ERβ LBD, and allosteric site. Our studies revealed that the
derivatives 3(a‐f) exhibited selectivity for ERα LBD. The
significant interacting residues of compounds were com-
pared with the interacting residues of 4HT. Compound 3d
was found to be most potent in the series with ‐8.71 docking
score and has exhibited significant interactions with key resi-
dues Arg394 and Glu353. Glu353 and Arg 394 are known to
establish H‐bonds with a 4‐hydroxyphenyl group of 4HT and
enhance complex stability [9]; similarly, in our case, we see
FIGURE 5 Transactivation analysis to show the effect of 3d
and 4HT in regulating ERα in MCF‐7 cells. The data are mean ± SEM
of three independent experiments performed in triplicates—one‐way
ANOVA where p‐value <0.05 is significant
Our initial attempt to generate nitrile oxides from aldoximes
using NBS in the presence of triethylamine in aprotic solvents
at 0ºC and further trapping them by dipolarophile‐2‐vinylchr-
omones resulted in a mixture of products which were difficult
to isolate using column chromatography. As an alternative
strategy, ceric ammonium nitrate (CAN) was used for gener-
ation of nitrile oxides (Das et al., 2005) from benzaldoximes
FIGURE 6 ERβ‐GFP Reporter
Assay: Fluorescence microscopic pictures
of GFP‐labeled ERβ in breast cancer cells
induced with E2, 4HT, and 3d following
2
4‐hr treatment. Cells on treatment with 3d
emitted enhanced fluorescence as compared
to control, E2, and 4HT

1
0
KAUSHIK et Al.
|
FIGURE 7 (a) Cell cycle analysis: Effect of 4HT and 3d in MCF‐7 cells. Cells were assessed for induction of apoptosis as seen by the
increase in the number of cells in sub-G0/G1 phase of cell cycle following 48‐hr treatment with HT (5, 10, 20 μg/ml) and 3d (15, 30, 60 μg/ml) by
propidium iodide staining. (b) Data are expressed as Mean±SD of apoptosis expressed as a percentage (n = 3). One‐way ANOVA where p‐value
<0.05 is considered significant
interactions of these residues with the phenol hydroxyl group
of compound 3d. The docking score of compounds 3(a–f) is
provided in Table 1.
The ligand interaction map of 3d exhibiting interactions
with ERα and the three‐dimensional structure of ERα and
ERβ docked with compound 3d are provided in Figure 2.
Comparison of compounds exhibiting interaction with resi-
dues in 3ERT is provided in Supporting information (Table
S3).
it was imperative to study 3d in detail to determine its recep-
tor selective nature. In this regard, other studies were also
carried out on 3d in MCF‐7 cells. We have also carried out
co‐treatment dose‐dependent and time‐dependent studies of
compound 3d with 4HT. Also, we have determined the antag-
onistic activity of compound 3d via modified E‐screen assay;
all results are provided in Supporting information (Figures
S1–S3).
4
.4
Gene silencing studies
|
4
.3
Dose‐dependent studies
|
The above results are indicative that compound 3d may indeed
be exhibiting ER‐mediated antagonistic behavior. To con-
firm the ERα selective nature of 3d, we have silenced ERα in
ERα+ MCF‐7 cells using specific siRNA against human ERα.
To confirm the silencing, the whole cell extract from silenced
and non‐silenced cells was prepared. The whole cell extract
was separated by electrophoresis, and Western blotting was
carried out as described in the Section 2. The specific anti-
body against ERα was used to detect ERα levels and compared
with the housekeeping protein β‐actin. As shown in Figure 4A,
ERα has been significantly down‐regulated in silenced cells
compared to control vector transfected MCF‐7 cells. The
control MCF‐7 and silenced cells were exposed to 4HT and
compound 3d for 48 hr followed by analysis of apoptosis by
Dose‐dependent studies were undertaken on compounds
3
(a–f) in ERα‐positive (MCF‐7) and ERα‐negative (MDA‐
MB‐231 and MRC‐5) cell lines as compared to controls
using MTT assay. Results indicated preferential cytotox-
icity of compound 3(a–f) in MCF‐7 cells. Compound 3d
was found to be most potent with IC value 31.25 μg/
5
0
ml. The IC value of 3(a–f) on ERα+ (MCF‐7) and ERα‐
5
0
(
MDA‐MB‐231) is given in Supporting information (Table
S2). The results of the dose‐dependent assay are given in
Figure 3.
Compound 3d was found to be a most potent cytotoxic
agent against ERα+ cells (MCF‐7). Moreover, docking stud-
ies with 3d indicated its selectivity for ERα LBD. Therefore,

KAUSHIK et Al.
11
|
FIGURE 8 Caspase activation
analysis: Effect of 4HT and 3d on triggering
caspase‐3 activation in MCF‐7 SCAT3 NLS
cells. Cells were assessed for cell death
following 24‐ and 48‐hr treatment with 4HT
(
5.63 μg/ml) and 3d (IC5 value = 32 μg/
0
ml) by FRET‐based assay. (a) Representative
images of merged and ratiometric panel
after 48 hr of treatment shown. (b) Data
are expressed as Mean±SD of cell death
expressed as a percentage (n = 3). Two‐way
ANOVA followed by Bonferroni post‐test
where p‐value <0.05 is significant
chromatin condensation as presented in Figure 4b, c. Silencing
of ERα reduced cell death induced by 4HT and 3d.
decrease in luciferase activity as compared to control at its
IC concentration. These data suggest that 3d‐induced re-
5
0
sponses are probably mediated through ERα antagonism.
4
.5
Luciferase reporter gene assay
|
4
.6
ERβ‐GFP Reporter Assay
|
The above results indicated anti‐estrogenic activity of tested
compound 3d. To further validate gene expression in the
presence of compound 3d, ERα‐positive MCF‐7 cells were
transiently transfected with PGL‐3‐(ERE)‐luciferase con-
struct. As depicted in Figure 5, compound 3d exhibited a
Above results predicted ERα‐dependent antagonism of com-
pound 3d. To further ascertain the ERα selective antagonistic
nature of compound 3d, we carried out ERβ‐GFP reporter
assay.

1
2
KAUSHIK et Al.
|
FIGURE 9 LC3 GFP assay: Effect
of 4HT‐1 (2.81 μg/ml) and 3d‐1(32 μg/
ml) on MCF‐7 cells expressing LC3
GFP and stained with TMRM (indicates
mitochondrial membrane potential).
Cells were assessed for autophagy and
mitochondrial membrane potential loss
following 24‐hr treatment with 4HT‐1
and 3d‐1
The results indicated a robust signal in the presence of com-
4
.9
LC3 GFP assay for autophagy and
|
pound 3d and estradiol (E2), while control cells elicited very
little response (fluorescence) as seen in Figure 6. These data
confirm that the same compound 3d has different effects on ERα
and ERβ; it is an antagonist to ERα while an agonist to ERβ.
mitochondrial membrane potential loss
Treating breast cancer cell line, MCF‐7 expressing LC3 GFP
failed to show a significant increase in LC3 punctae in 3d‐
treated cells compared to untreated cells. However, 4HT at
2
.81 μg/ml significantly induced LC3 punctae in treated cells.
4
.7
Cell cycle analysis by Propidium
|
The percentage of cells with more than 15 punctae per cell
scored for 4HT, and 3d is shown in Figure 9. Interestingly, both
3d and 4HT induced an almost similar variation in mitochon-
drial membrane potential (as evident from TMRM staining) in
treated cell. The result suggests that 3d is a dominant cell death
inducer with minimum autophagy; however, 4HT simultane-
ously induced massive survival signaling through the induction
of autophagy.
iodide Staining
The effect of compound 3d on the cell cycle distribution and
sub‐G0/G1 phase on MCF‐7 cells was measured to determine
whether cytotoxicity on ERα is mediated via apoptosis and
cell cycle modulation. Treatment with 3d at 30 and 60 μg/ml
induced increase in the number of cells in sub‐G0/G1 level
as compared to control in MCF‐7 cells. This indicates com-
pound 3d induced cell death via apoptosis similar to 4HT
(
Figure 7).
4
.10
Increase in ROS contributed to cell
|
death induced by 3d
4
.8
Cell death pathway
|
MCF‐7 cells expressing redox‐sensitive GFP was employed
to analyze the levels of intracellular ROS induced by 3d.
The redox GFP is a reduction–oxidation sensitive engi-
neered GFP probe that interacts with glutaredoxin and acts
as a sensor of cellular redox potential based on the variation
in its GFP emission at two distinct excitation wavelengths
405 nm and 488 nm. The cells were exposed to 3d and 4HT
for 24 hr followed by ratio analysis, both by flow cytom-
etry and by microscopy. As shown in Figure 10, treatment
of 3d induced a significant increase in cells with enhanced
405/488 ratios by both flow cytometry and microscopy. 4HT
also induced an increase in 405/488 ratio in the majority of
The chromatin condensation results (Figure 4) and cell cycle
analysis (Figure 7) indicated that cell death induced by com-
pound 3d involved apoptosis. Further to confirm pathway
of apoptosis, a sensitive live cell caspase activation analysis
using ECFP/EYFP‐based FRET probe expressing cells was
employed. In MCF‐7 cells expressing SCAT3 NLS, caspase
detection FRET probe was exposed to compound 3d, as de-
scribed in Materials and Methods. The result indicated that
compound 3d triggered activation of the caspase‐3‐mediated
cell death pathway with a maximal increase at 48 hr of treat-
ment (Figure 8).

KAUSHIK et Al.
13
|
FIGURE 10 (a) ROS analysis: Effect of 4HT and 3d in MCF‐7 cells expressing ratiometric ROS sensing probe, roGFP. Cells were assessed
for change in ROS level following 24‐hr treatment with HT (HT‐1 = 5.63 μg/ml and HT‐2 = 9.7 μg/ml) and 3d (3d‐1 = 32 μg/ml and 3d‐2 = 62 μg/
ml) by ratiometric analysis using FACS. (b) Data are expressed as Mean±SD of cell death expressed as a percentage (n = 3). Two‐way ANOVA
followed by Bonferroni post‐test where p‐value <0.05 is significant. (c) The effect of 4HT (4HT‐1) and 3d (3d‐1) on ROS level was reinforced from
the ratiometric images after 24 hr of treatment
cells suggesting ROS as the contributing factor for its cell
death.
2013). Even though a majority of them proved effective
in clinical settings, a subset of breast cancer shows resist-
ance to SERM owing to diverse signaling defects of ERα.
In addition, several of the compounds suffer from severe
off‐target effects such as endocrine resistance, osteoporosis,
hot flashes, deep vein thrombosis (DVT), pulmonary embo-
lism, and endometrial cancer (Ellis, Hendrick, Williams, &
Komm, 2015; Maximov et al., 2013). Hence, increased atten-
tion has been directed toward identifying novel SERM with
less off‐target effects based on both natural product screen-
ing and employing synthetic chemistry. The development of
5
DISCUSSION
|
Estrogen receptor signaling plays a critical role in driving
ER‐positive breast cancer and considered as an important
druggable target. Tamoxifen along with second‐ and third‐
generation SERMs are currently used in clinical setting for
the treatment of ER‐positive breast cancer (Maximov et al.,

1
4
KAUSHIK et Al.
|
subtype‐selective ligands that specifically target either ERα
or ERβ would be a more optimal approach for the treatment
of cancer and other diseases such as cardiovascular disease,
multiple sclerosis, and Alzheimer's (Nilsson, Koehler, &
Gustafsson, 2011). Diverse efforts have been made toward
the synthesis of newer ER modulators with increased selec-
tivity toward ERα‐positive breast cancer cells and with fewer
side‐effects. Beside steroid hormones, ER targeting scaf-
folds include flavones, isoflavones, and coumestans as well
as diphenylethylenic derivatives and analogues in the trans
configuration (Leclercq, Lacroix, Laios, & Laurent, 2006).
Grafting of hydrophobic substituents (i.e., phenyl groups) at
specific locations in type I estrogens may transform agonists
into type II antagonists (Leclercq et al., 2006). Polycyclic
compounds containing a phenolic ring and other oxygen-
ated heterocycles located at the opposite end of the molecule
hormone‐dependent cells involving ERα (Huang, Warner,
& Gustafsson, 2015). In addition compared to 4HT, 3d
has failed to induce significant survival signaling in the
treated cells involving autophagy. Induction of autoph-
agy is one of the key determinants of hormone resistance
(Cook, Shajahan, & Clarke, 2011; Schoenlein, Periyasamy‐
Thandavan, Samaddar, Jackson, & Barrett, 2009), and sev-
eral of the clinically approved SERM promote autophagy
activity (Samaddar et al., 2008; Viedma‐Rodríguez et al.,
2014) . Furthermore, we provide evidence that 3d is the
most effective compound that primarily induces ROS‐de-
pendent cell death through mitochondrial membrane poten-
tial loss and caspase activation. This class of compound
is expected to provide an additional class of structurally
different new SERM for further development and testing in
preclinical models in the best interest of the development
of new SERM.
3
where the distance between oxygen is ~11 Å have been
reported to target ER (Leclercq et al., 2006). Large internal
hydrocarbon moiety in these molecules contributes to an
optimal orientation of polar functions for selective H‐bond-
ing with specific amino acid residues of the ligand‐binding
pocket (LBD) of ER. Linear, planar molecules are usually ER
agonists whereas angular geometry leads to anti‐estrogenic-
ity (Leclercq et al., 2006). Reports also suggest molecules
that exhibit planar topology and are conformationally rigid
usually exhibit ERβ selectivity. Hsieh et al., 2006 reported
that ERβ selectivity of oxabicyclic ligands is attributed to the
close interaction of ligands with Met‐336(ERβ) residue lead-
ing to ERβ subtype selectivity. Further, it was also observed
that unfavorable interaction with Met421 in ERα resulted in
ligands exhibiting ERβ selectivity prominently.
In summary, we report synthesis and validation of a new
class of isoxazolylchromones 3(a–f) as a novel ERα selec-
tive agent with promising bioactivity based on a different
structural design, as evident from bioactivity profiling,
selectivity studies, and comparative in vitro cytotoxicity
results. The results confirm its utility as an ER isoform se-
lective agent with the potential to be developed as possible
SERMs. Further studies are ongoing to substantiate their
off‐target effects and clinical potential in patient‐derived
samples.
ACKNOWLEDGMENTS
SK thanks SERB‐NPDF (PDF/2016/004001) for finan-
cial assistance. All authors gratefully acknowledge Prof.
M. Radhakrishna Pillai, Director, Rajiv Gandhi Centre
for Biotechnology, Trivandrum, for his support. Authors
also gratefully acknowledge the guidance of Dr.Damoder
Gupta, INMAS, for cytotoxicity studies. Authors acknowl-
edge Amity Institute of Biotechnology, Amity University,
for infrastructure and other facilities. Support to the Cancer
ResearchProgram1, RajivGandhiCentreforBiotechnology,
Trivandrum, by Department of Biotechnology (BT/
PR8087/MED/32/292/2013), Government of India, is also
gratefully acknowledged. SK also acknowledges Prakash
R. Pillai, Technical Assistance, Rajiv Gandhi Centre for
Biotechnology, Trivandrum, for his timely assistance.
Here, we report the development of isoxazolylchromones
as a new class of SERM with strong bioactivity and selec-
tivity toward ERα. The initial docking results indicated that
compounds did not exhibit interaction with key residue Met
3
36 in ERβ; however, all ligands exhibited interaction with
Met 421 in ERα. Moreover, several common interactions
were found as compared to 4HT in ERα most notably Arg
3
94. Accordingly, ERα luciferase activity confirmed the an-
tagonistic activity of compound 3d in relevant breast cancer
cell model, MCF‐7. The experiments utilizing FRET caspase
probe and FACS analysis further confirmed the cytotoxicity
and ERα antagonistic potential of the compound in relation to
the clinically used SERM, 4HT.
Since we have employed a new approach of chromone
scaffold as the building block to synthesize ERα modu-
lators, it has exhibited strong bioactivity and better ERα
binding in all the biological assays. A promising feature
of 3d seems to be its ability to induce significant ERβ
transactivation. Previous studies have demonstrated that
ERα inhibition induces ERβ as a feedback signaling (Saji,
Hirose, & Toi, 2005) and ERβ transactivation is known to
be advantageous in inhibiting the proliferation potential of
CONFLICT OF INTEREST
The author(s) declare no competing interests.
ORCID
Thankayyan Retnabai Santhoshkumar
org/0000-0003-3386-7105
https://orcid.

KAUSHIK et Al.
15
|
REFERENCES
transcriptional activity in MCF‐7 cells. Biochemical Pharmacology,
7
4, 1029–38. https://doi.org/10.1016/j.bcp.2007.06.045
Banks, J. L., Beard, H. S., Cao, Y., Cho, A. E., Damm, W., Farid, R., …
Levy, R. M. (2005). Integrated modeling program, applied chemical
theory (IMPACT). Journal of Computational Chemistry, 26, 1752–
Leclercq, G., Lacroix, M., Laios, I., & Laurent, G. (2006). Estrogen re-
ceptor alpha: Impact of ligands on intracellular shuttling and turn-
over rate in breast cancer cells. Current Cancer Drug Targets, 6,
1
780. https://doi.org/10.1002/jcc.20292
3
9–64. https://doi.org/10.2174/156800906775471716
Beato, M., Herrlich, P., & Schütz, G. (1995). Steroid hormone recep-
tors: Many actors in search of a plot. Cell, 83, 851–857. https://doi.
org/10.1016/0092-8674(95)90201-5
Masood, S. (1992). Estrogen and progesterone receptors in cytology:
A comprehensive review. Diagnostic Cytopathology, 8, 475–491.
https://doi.org/10.1002/dc.2840080508
Maximov, P. Y., Lee, T. M., & Jordan, V. C. (2013). The discovery and
development of selective estrogen receptor modulators (SERMs)
for clinical practice. Current Clinical Pharmacology, 8, 135–155.
https://doi.org/10.2174/1574884711308020006
Middleton, E. Jr., Kandaswami, C., & Theoharides, T. C. (2000). The
effects of plant flavonoids on mammalian cells: Implications for in-
flammation, heart disease, and cancer. Pharmacological Reviews,
Chang, B. Y., Kim, S. A., Malla, B., & Kim, S. Y. (2011). The effect of
selective estrogen receptor modulators (SERMs) on the tamoxifen
resistant breast cancer cells. Toxicology Research, 27, 85–93. https://
doi.org/10.5487/TR.2011.27.2.085
Cook, K. L., Shajahan, A. N., & Clarke, R. (2011). Autophagy and en-
docrine resistance in breast cancer. Expert Review of Anticancer
Therapy, 11, 1283–1294. https://doi.org/10.1586/era.11.111
Das, B., Mahender, G., Holla, H., & Banerjee, J. (2005). An easy access
to 3a,4‐dihydro‐3H‐chromeno[4,3‐c]isoxazoles and functionalized
isoxazolines. ARKIVOC, 3, 27–35.
Ellis, A. J., Hendrick, V. M., Williams, R., & Komm, B. S. (2015).
Selective estrogen receptor modulators in clinical practice: A safety
overview. Expert Opinion in Drug Safety, 14, 921–934. https://doi.
org/10.1517/14740338.2015.1014799
Farhat, M. Y., Lavigne, M. C., & Ramwell, P. W. (1996). The vascular
protective effects of estrogen. The FASEB Journal, 10, 615–624.
Heldring, N., Pike, A., Andersson, S., Mathews, J., Cheng, G., Hartman,
J., … Gustafsson, J. A. (2007). Estrogen receptors: How do they sig-
nal and what are their targets. Physiological Reviews, 87, 905–931.
https://doi.org/10.1152/physrev.00026.2006
Henderson, B. E., & Feigelson, H. S. (2000). Hormonal carcino-
genesis. Carcinogenesis, 21, 427–433. https://doi.org/10.1093/
carcin/22.3.529
Hongyi, Q., Zhuyun, J., Chengqiang, W., Yang, Y., Li, L., Hui, H., &
Zanyang, Y. (2017). Sensitization of tamoxifen‐resistant breast can-
cer cells by Z‐ligustilide through inhibiting autophagy and accu-
mulating DNA damages. Oncotarget, 8, 29300–29317. https://doi.
org/10.18632/oncotarget.16832
5
2, 673–751.
Minutolo,F.,Macchia,M.,Katzenellenbogen,B.S.,&Katzenellenbogen,
J. A. (2011). Estrogen receptor β ligands: Recent advances and bio-
medical applications. Medicinal Research Reviews, 31, 364–442.
https://doi.org/10.1002/med.20186
Momoi, K., Sugita, Y., Ishihara, M., Satoh, K., Kikuchi, H., Hashimoto,
K., … Sakagami, H. (2005). Cytotoxic activity of styrylchromones
against human tumor cell lines. In Vivo, 19, 157–163.
Nilsson, S., Koehler, K. F., & Gustafsson, J. A. (2011). Development
of subtype‐selective oestrogen receptor‐based therapeutics. Nature
Reviews Drug Discovery, 10, 778–792. https://doi.org/10.1038/
nrd3551
Nilsson, S., Makela, S., Treuter, E., Tujague, M., Thomsen, J.,
Andersson, G., … Gustafsson, J. A. (2001). Mechanism of estro-
gen action. Physiological Reviews, 81, 1535–1565. https://doi.
org/10.1152/physrev.2001.81.4.1535
Ottow, E., & Weinmann, H. (2008). Nuclear receptors as drug targets:
A historical perspective of modern drug discovery. New York, NY:
Wiley‐VCH Verlag GmbH & Co. KGaA, Wiley Publications.
Paterni, I., Granchi, C., Katzenellenbogen, J. A., & Minutolo, F. (2014).
Estrogen receptors alpha (ERα) and beta (ERβ): Subtype‐selec-
tive ligands and clinical potential. Steroids, 90, 13–29. https://doi.
org/10.1016/j.steroids.2014.06.012
Hsieh, R. W., Rajan, S. S., Sharma, S. K., Guo, Y., DeSombre, E. R.,
Mrksich, M., & Greene, G. L. (2006). Identification of ligands with
bicyclic scaffolds provides insights into mechanisms of estrogen
receptor subtype selectivity. The Journal of Biological Chemistry,
Peng, J., Sengupta, S., & Jordan, V. C. (2009). Potential of selective
estrogen receptor modulators as treatments and preventives of breast
cancer. Anti‐cancer Agents in Medicinal Chemistry, 9, 481–99.
Pick, A., Müller, H., Mayer, R., Haenisch, B., Pajeva, I. K., Weigt, M.,
2
81, 17909–17919. https://doi.org/10.1074/jbc.M513684200
Huang, B., Warner, M., & Gustafsson, J. A. (2015). Estrogen recep-
tors in breast carcinogenesis and endocrine therapy. Molecular and
Cellular Endocrinology, 418, 240–444. https://doi.org/10.1016/j.
mce.2014.11.015
…
Wiese, M. (2011). Structure‐activity relationships of flavonoids
as inhibitors of breast cancer resistance protein (BCRP). Bioorganic
&
Medicinal Chemistry, 19, 2090–2102. https://doi.org/10.1016/j.
Joseph, J., Seervi, M., Sobhan, P. K., & Santhoshkumar, T. R. (2011).
High throughput ratio imaging to profile caspase activity: Potential
application in multiparameter high content apoptosis analysis and
drug screening. PLoS ONE, 6, 1–17. https://doi.org/10.1371/jour-
nal.pone.0020114
Kuiper, G. G., Carlsson, B., Grandien, K., Enmark, E., Häggblad, J.,
Nilsson, S., & Gustafsson, J. A. (1997). Comparison of the ligand
binding specificity and transcript tissue distribution of estrogen
receptors alpha and beta. Endocrinology, 138, 863–70. https://doi.
org/10.1210/endo.138.3.4979
bmc.2010.12.043
Richardson, T. I., Norman, B. H., Lugar, C. W., Jones, S. A., Wang,
Y., Durbin, J. D., … Dodge, J. A. (2007). Benzopyrans as selective
estrogen receptor beta agonists (SERBAs). Part 2: Structure‐ac-
tivity relationship studies on the benzopyran scaffold. Bioorganic
&
Medicinal Chemistry Letters, 17, 3570–3574. https://doi.
org/10.1016/j.bmcl.2007.04.051
Saji, S., Hirose, M., & Toi, M. (2005). Clinical significance of es-
trogen receptor beta in breast cancer. Cancer Chemotherapy
and Pharmacology, 56, 21–26. https://doi.org/10.1007/
s00280-005-0107-3
Laïos, I., Cleeren, A., Leclercq, G., Nonclercq, D., Laurent, G., Schlenk,
M., …Gust, R. (2007). Effectsof(R, S)/(S, R)‐4,5‐bis(2‐chloro‐4‐hy-
droxyphenyl)‐2‐imidazolines and (R, S)/(S, R)‐2,3‐bis(2‐chloro‐4‐
hydroxyphenyl) piperazines on estrogen receptor alpha level and
Samaddar, J. S., Gaddy, V. T., Duplantier, J., Thandavan, S. P., Shah, M.,
Smith, M. J., … Schoenlein, P. V. (2008). A role for macroautophagy

1
6
KAUSHIK et Al.
|
in protection against 4‐hydroxytamoxifen‐induced cell death and
the development of antiestrogen resistance. Molecular Cancer
Therapeutics, 7, 2977–2987. https://doi.org/10.1158/1535-7163.
MCT-08-0447
resistance to tamoxifen in estrogen receptor‐positive breast cancer.
Oncology Reports, 32, 3–15. https://doi.org/10.3892/or.2014.3190
Wang, Y., Chirgadze, N. Y., Briggs, S. L., Khan, S., Jensen, E. V., &
Burris, T. P. (2006). A second binding site for hydroxytamoxifen
within the coactivator‐binding groove of estrogen receptor beta.
Proceedings of the National Academy of Sciences USA, 103, 9908–
9911. https://doi.org/10.1073/pnas.0510596103
Schoenlein, P. V., Periyasamy‐Thandavan, S., Samaddar, J. S., Jackson,
W. H., & Barrett, J. T. (2009). Autophagy facilitates the progression
of ER alpha‐positive breast cancer cells to antiestrogen resistance.
Autophagy, 5, 400–403. https://doi.org/10.4161/auto.5.3.7784
Shiau, A. K., Barstad, D., Loria, P. M., Cheng, L., Kushner, P. J.,
Agard, D. A., & Greene, G. L. (1998). The structural basis of es-
trogen receptor/coactivator recognition and the antagonism of this
interaction by tamoxifen. Cell, 95, 927–37. https://doi.org/10.1016/
S0092-8674(00)81717-1
SUPPORTING INFORMATION
Additional supporting information may be found online in
the Supporting Information section at the end of the article.
Sommer, S., & Fuqua, S. A. (2001). Estrogen receptor and breast cancer.
Seminars in Cancer Biology, 11, 339–352. https://doi.org/10.1006/
scbi.2001.0389
Turner, R. T., Riggs, B. L., & Spelsberg, T. C. (1994). Skeletal effects of
estrogen. Endocrine Reviews, 15, 275–300. https://doi.org/10.1210/
edrv-15-3-275
Viedma‐Rodríguez, R., Baiza‐Gutman, L., Salamanca‐Gómez, F., Diaz‐
Zaragoza, M., Martínez‐Hernández, G., Ruiz Esparza‐Garrido,
R., … Arenas‐Aranda, D. (2014). Mechanisms associated with
How to cite this article: Kaushik S, Sanawar R,
Lekshmi A, et al. ER alpha selective chromone,
isoxazolylchromones, induces ROS‐mediated cell
death without autophagy. Chem Biol Drug Des.
2
019;00:1–16. https://doi.org/10.1111/cbdd.13510